Here's the pattern for a production-ready chatbot API using OpenAI and FastAPI:

from fastapi import FastAPI
from openai import OpenAI
from pydantic import BaseModel

app = FastAPI()
client = OpenAI()

class ChatRequest(BaseModel):
    message: str
    history: list = []

@app.post("/chat")
def chat(req: ChatRequest):
    messages = [
        {"role": "system", "content": "You are a helpful assistant."},
        *req.history,
        {"role": "user", "content": req.message}
    ]
    response = client.chat.completions.create(
        model="gpt-4o", messages=messages
    )
    return {"reply": response.choices[0].message.content}

Reference:

Production best practices

https://en.wikipedia.org/wiki/Special:Search?search=Building%20a%20Chatbot